Git.Top Benchmark API
The Benchmark API from Git.Top — 1 operation(s) for benchmark.
Work with this as data
Every API here is available over the APIs.io API and to AI agents over MCP.
MCP server
One button, every client — Claude, Cursor, VS Code and the rest.
https://apis.io/mcp
Tools for apis
7 MCP tools reach this
find_apisBrowse and filter every API in the catalog.get_api_artifactsOne API's artifacts, grouped by type.get_openapiThe primary OpenAPI for this API.find_similar_apisAPIs that look like this one.apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.resolveTurn a domain, URL or GitHub org into the provider it belongs to.find_cohortsEvery scored population of providers in the catalog.
Call it yourself
curl for this page
This API
curl "https://apis.io/api/v1/apis/git-top:git-top-benchmark-api"
All apis
curl "https://apis.io/api/v1/apis?limit=25"Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.
A second provider on the same verified email joins the account you already have.
OpenAPI Specification
openapi: 3.2.0
info:
title: Git.Top Agent Benchmark API
version: 0.1.0
description: Agent-native GitHub project knowledge API for search, project lookup, alternatives, comparison, graph reasoning, quality, and schema discovery. High-confidence clients should inspect metadata.source, metadata.reason, trust fields, classification evidence, quality_signal_confidence, verification signals, and migration validation steps, and use require_d1=true when seed fallback should fail closed.
servers:
- url: https://git.top
description: Production
- url: http://localhost:8787
description: Local Worker
tags:
- name: Benchmark
paths:
/api/benchmark:
get:
summary: Inspect public recommendation, explanation, data trust, review queue, and…
parameters:
- name: require_d1
in: query
required: false
description: Fail closed unless D1-backed data is available
schema:
type: string
responses:
'200':
description: Public trust benchmark summary with eval health, explanation coverage, data coverage, review queue, and known limitations
content:
application/json:
schema:
$ref: '#/components/schemas/BenchmarkResponse'
example:
name: Git.Top Public Trust Benchmark
positioning: The Knowledge Graph of Open Source
summary: Recommendation eval top-3 hit rate is 1; explanation checks are 14/14; data trust is 82/100 with 12 review queue items.
generated_at: '2026-07-04T00:00:00.000Z'
evaluation:
generated_at: '2026-06-30T00:08:13.734Z'
evaluated_cases: 28
top1_hit_rate: 0.929
top3_hit_rate: 1
category_accuracy: 1
deployment_accuracy: 1
cloudflare_readiness_accuracy: 1
unacceptable_hit_count: 0
review_focus:
- id: search-coding-agent
issue: top-1 miss
expected:
- openai/codex
observed:
- smol-ai/developer
source_report: docs/EVAL_QUALITY.md
explanations:
generated_at: '2026-07-04T02:51:39.580Z'
checks: 14
passed: 14
failed: 0
coverage: 1
source_report: docs/EVAL_EXPLANATIONS.md
data_coverage:
project_count: 500
category_coverage: 14/14
low_confidence_classification_rate: 0.04
data_trust_score: 82
release_score: 96
risk_level: low
review_queue:
review_count: 12
low_signal_count: 18
top_impact_score: 88
top_items:
- project_id: cloudflare/agents
category: agent_framework
impact_score: 88
known_limitations:
- Eval quality is a CI-safe baseline over curated cases and generated fixtures.
recommended_use:
- Use top_3_hit_rate, explanation coverage, data_trust_score, and review_queue before high-confidence production recommendations.
links:
html: /benchmark
api: /api/benchmark
quality: /api/quality
review: /api/quality/review
trust: /api/trust
metadata:
source: d1
reason: d1_query
project_count: 500
generated_at: '2026-07-04T00:00:00.000Z'
snapshot_id: d1:500:2026-07-04T00:00:00.000Z
latest_synced_at: '2026-07-04T00:00:00.000Z'
schema_version: git-top.knowledge.v1
tags:
- Benchmark
operationId: getApiBenchmark
x-operation-id-source: derived
components:
schemas:
BenchmarkResponse:
type: object
description: Public benchmark summary for agents and users evaluating Git.Top recommendation health and trust limits.
required:
- name
- evaluation
- explanations
- data_coverage
- review_queue
- known_limitations
- metadata
properties:
name:
type: string
positioning:
type: string
summary:
type: string
generated_at:
type: string
format: date-time
evaluation:
type: object
properties:
top1_hit_rate:
type: number
top3_hit_rate:
type: number
category_accuracy:
type: number
deployment_accuracy:
type: number
cloudflare_readiness_accuracy:
type: number
unacceptable_hit_count:
type: integer
review_focus:
type: array
items:
type: object
additionalProperties: true
source_report:
type: string
additionalProperties: true
explanations:
type: object
properties:
checks:
type: integer
passed:
type: integer
failed:
type: integer
coverage:
type: number
source_report:
type: string
additionalProperties: true
data_coverage:
type: object
additionalProperties: true
review_queue:
type: object
additionalProperties: true
known_limitations:
type: array
items:
type: string
recommended_use:
type: array
items:
type: string
links:
type: object
additionalProperties:
type: string
metadata:
$ref: '#/components/schemas/Metadata'
additionalProperties: true
Metadata:
type: object
description: Common provenance envelope for agent-facing responses. Agents should require source=d1 for high-confidence production claims and disclose seed fallback.
properties:
source:
type: string
enum:
- d1
- seed
reason:
type: string
project_count:
type: integer
generated_at:
type: string
format: date-time
snapshot_id:
type: string
description: Stable identity for the corpus snapshot used by this response.
latest_synced_at:
type:
- string
- 'null'
format: date-time
schema_version:
type: string
enum:
- git-top.knowledge.v1
loaded_project_limit:
type: integer
description: Maximum D1 knowledge or search-candidate rows loaded into the bounded ranking set when present.
truncated:
type: boolean
description: Whether the D1-backed ranking or candidate set reached its limit and may omit additional matching rows.
candidate_retrieval:
type: string
enum:
- d1_first
description: Search-only marker indicating that deterministic filters and broad query matching ran in D1 before TypeScript ranking.
candidate_count:
type: integer
description: Number of D1 search candidates passed to the authoritative TypeScript filter and ranking stage.
candidate_limit:
type: integer
description: Maximum number of D1 search candidates passed to the ranking stage.
warnings:
type: array
items:
type: string
description: Fallback, count, freshness, or trust caveats that agents should account for before citing results.
securitySchemes:
syncSecret:
type: http
scheme: bearer
description: Admin endpoints require the SYNC_SECRET bearer token.
feedbackSecret:
type: http
scheme: bearer
description: Optional FEEDBACK_SECRET. Without it, proposals are validated and fingerprinted but not persisted.